The discrete element method (DEM) is widely used as an optimization tool in the design of tumbling mills. Nevertheless, the experimental validation of DEM codes is an important step to guarantee that the system is well described and that the predictions are close to the real operating conditions. The power draw is the principal parameter used to validate DEM codes, but using this as the only reference can be inappropriate. The power draw can be fitted by simply changing the model constants until the predicted values are close to the experimental data, when the charge profile looks similar to the real operation and there is no experimental velocity profile with which to compare. This paper presents an experimental validation of a 20 DEM code by digital image analysis of the velocity profiles of the balls, the toe and shoulder angles and the predicted power draw. The experimental values were compared with the simulated data using different charge lifters and charge levels. The DEM simulation clearly shows that the velocity charge changes with a modification of the lifter profile. An accurate map of velocities at each location in the mill was obtained by digital image analysis and compared with DEM calculations. The simulated and experimental values are very close, leading to the conclusion that such DEM predictions represent an accurate description of the process in a tumbling mill. (C) 2011 Elsevier Ltd. All rights reserved.